i?,i:CEIVE!!
BEFORE THE
POSTAL RATE COMMISSION
WASHINGTON. DC. 20266-0001
POSTAL RATE AND FEE CHANGES, 2000
/
o:q(,;: I’,i #,Z,>L!/
Docket No. R2006-1
1
RESPONSE OF UNITED STATES POSTAL SERVICE
WITNESS YEZER TO INTERROGATORIES OF
THE OFFICE OF THE CONSUMER ADVOCATE
(OCAIUSPS-T31-1-5)
The United States Postal Service hereby provides the responses of witness
Yezer to the fotlowing interrogatories of the Oftice of the Consumer Advocate:
.
OCAIUSPS-T31-I-5, filed on February 14.2000. .~
Each interrogatory is stated verbatim and is followed by the response.
Respectfully submitted,
UNITED STATES POSTAL SERVICE
By its attorneys:
Daniel J. Foucheaux, Jr.
Chief Counsel, Ratemaking
kl, $JL
Kenneth N. Hollies
RESPONSE OF UNITED STATES POSTAL SERVICE WITNESS YEZER TO
INTERROGATORIES OF OFFICE OF THE CONSUMER ADVOCATE
OCA/USPS-T31-1. Please refer to your testimony at page 2. lines 7-l 1.
(a)
Please identify by name and provide citations to the “significantly smaller
group of papers” that have considered office rents in selected cities.
(b)
Please explain how the predictive model developed in your research for
the Postal Service is consistent with the research in the papers identified in part
(a) of this interrogatory.
(c)
Please explain how the predictive model developed in your research for
the Postal Service is inconsistent with the research in the papers identified in part
(a) of this interrogatory.
RESPONSE:
(4
(1) William C. Wheaton, Raymond G. Torto, and Jon A. Southard, “The
CB CommerciakTorto Wheaton Database,” Journal of Real Estate
Literature, Volume 5. 1997.
..~ Pages 59-139.
.~’
.:
(2) William C. Wheaton and Raymond G. Totto. “Office Rent Indexes and
Their Behavior Over Time,” Journal of Urban Economics, Volume 35.
1994. Pages 121-139.
(3) Brian Brennan, Roger Cannaday, and Peter Colwell, “Office Rent in
the Chicago CBD,” AREUEA Journal, Volume 12.1964. Pages 243-260.
04
Following the literature, I use data on individual leases and explain rent
per square foot of space in terms of lease terms, physical characteristics
of the property including interior space, and location.
(cl
My estimates cover the entire country, as opposed to one city or a group
of 56 cities, and model rents within cities in terms of distances from the
central CBD measured precisely in miles as opposed to rents in
predetermined city districts. I use rent per square foot rather than the
logarithm of rent per square foot as a dependent variable. My rent data is
all from a given renter, the Postal Service, as opposed to heterogeneous
R2000-1
RESPONSE OF UNITED STATES POSTAL SERVICE WITNESS YEZER TO
INTERROGATORIES OF OFFICE OF THE CONSUMER ADVOCATE
renters in other studies. I have more information on property
characteristics and lease terms than others have had and this means that
my model has more independent variables. I use robust estimation
techniques to deal with measurement error and I strongly suspect that
other authors use casewise deletion.
R2000-1
RESPONSE OF UNITED STATES POSTAL SERVICE WITNESS YEZER TO
INTERROGATORIES OF OFFICE OF THE CONSUMER ADVOCATE
OCAfUSPS-T31-2. Please refer to your testimony at pages 2 and 3 lines 20-21
and lines 1-2, respectively. Please explain whether you agree with the Postal
Service’s decision to use the econometric model developed in your research to
predict rent per square foot for specific facilities rather than an index of rental
price differences over space. In your answer, please compare and contrast the
advantages and disadvantages of the former and latter.
RESPONSE:
I was commissioned as an expert witness because of my knowledge of
real estate economics and not as an expert on the regulatory process faced by
the Postal Service. Nevertheless, the difference in using facility-specific rental
estimates versus a rental price index appears straightforward. Use of facilityspecific rental estimates means that post office box costs will reflect both spatial
variation in rental prices and rent variation due to the quality of the specific space
in the facility. For example, space in a shopping mall is generally significantly
more expensive than space in general retail facilities. Should the extra
convenience and extra cost associated with location in malls and office buildings
be reflected in P.O. box pricing? It seems to me that this is appropriate.
R2000-1
RESPONSE OF UNITED STATES POSTAL SERVICE WITNESS YEZER TO
INTERROGATORIES OF OFFICE OF THE CONSUMER ADVOCATE
OCA/USPS-T31-3. Please refer to your testimony at page 3, lines 9-10. Has
your research for the Postal Service been used to predict rent per square foot at
facilities that do not provide post office box service?
RESPONSE:
In the course of my research, I estimated rents for all Postal Service
facilities in the data base given to me and these estimates were forwarded to
Postal Service staff. In cases where there was more than one facility per ZIP
Code, I reported estimated rent for the facility with the most post ofke boxes.
R2000-1
RESPONSE OF UNITED STATES POSTAL SERVICE WITNESS YEZER TO
INTERROGATORIES OF OFFICE OF THE CONSUMER ADVOCATE
OCA/USPS-T31-4. Please refer to your testimony at page 3, lines 1 l-l 7.
Please confirm that the Postal Service offers post office box service in
(4
facilities that are postal-owned. If you do not confirm, please explain.
(b)
Please explain how your econometric model predicts the rent per square
foot in postal-owned facilities having post office boxes. In your answer,
please explain any assumptions about the age, location, condition, and
suitability of postal-owned facilities having post office boxes as compared
to leased facilities having post office boxes.
RESPONSE:
(a)
Many facilities that appear to be postal-owned in my database have post
oflice boxes.
(b)
My econometric model is used to predict rents in all facilities. whether they
are rented or postal-owned. The data do not include information on the
age of buildings or the condition. I am,not sure what the term suitability
means in this context. Given the property address information, I use
geographic information systems to determine location precisely. There is
also information about setting of the facility that differentiates shopping
malls, office parks, free standing retail buildings, etc.
Fuooo-1
RESPONSE OF UNITED STATES POSTAL SERVICE WITNESS YEZER TO
INTERROGATORIES OF OFFICE OF THE CONSUMER ADVOCATE
OCAIUSPS-T31-5. Please refer to your testimony at pages 6 and 7, lines 1920, and line 1 respectively. Please explain how knowing the “distance northsouth or east-west from the [Central Business District)” improves the prediction of
rent per square foot.
RESPONSE: In some cities the spatial pattern of ofice rents looks like a large
cone with the point at the center of the Central Business District (CBD). In such
cases, distance north-south or east-west was not statistically significant and was
dropped from the list of independent variables used. However, the pattern in
other cities looks like a cone but the point of the cone is displaced from the CBD
center. In such cases, either (or both) north-south or east-west distance was
significant in the regression estimates and was included in the final estimating
equation. In a few cities, particularly on the coast, distance east-west is very
important because it reflects access to the coast.
R2000-1
DECLARATION
I, Anthony M. Yezer, declare under penalty of perjury that the
foregoing answers are true and correct, to the best of my knowledge,
information, and belief.
Dated:
.11;1<lo2
CERTIFICATE OF SERVICE
I hereby certify that I have this day served the foregoing document upon all
participants of record in this proceeding in accordance with section 12 of the Rules of
Practice.
475 L’Enfant Plaza West, S.W.
Washington, D.C. 20260-l 137
(202) 266-3063 Fax -5402
February 252000